1. Field of the Invention
The present invention relates generally to an apparatus and method for determining a power control scheme in a time division duplex (TDD) mobile communication system, and in particular, to an apparatus and method for changing an uplink power control scheme according to the status of a subscriber station (SS).
2. Description of the Related Art
As one of the duplex schemes, Time Division Duplex (TDD) uses two distinct sets of time slots on the same frequency for the uplink from a base station (BS) to a Subscriber Station (SS) and the downlink from the SS to the BS. Another major duplex scheme is frequency division duplex (FDD). FDD uses two distinct frequencies for the uplink and the downlink.
Unlike FDD, the uplink and the downlink share the same frequency band in TDD and are separated by time slots dedicated to them. That is, time slots are separately preset for the uplink signal and the downlink signal. Therefore, the uplink and downlink signals are transmitted only in their assigned time slots. TDD has the advantage of high frequency use efficiency.
The mobile communication system schedules bursty uplink/downlink packets. Particularly, the BS decides a modulation and coding scheme (MCS) for the resources to be allocated and already allocated resources in uplink/downlink packet scheduling for an SS. An MCS level to be used depends on the status of the SS. For the uplink scheduling, the BS takes into account the maximum transmit power of the SS. Since the transmit power of the SS is restricted to a set level, the BS performs scheduling taking into account the allocated resources, an MCS level to be applied for the resources, and the transmit power limit of the SS. To do so, the scheduler of the BS must have knowledge of the power headroom or transmit power of the SS.
Typically, the mobile communication system uses downlink and uplink power control to increase call capacity and achieve good call quality. That is, if the BS receives a signal from an SS at a signal-to-interference ratio (SIR) that ensures the minimum required call quality by controlling the transmit power of all of the SSs, system capacity can be maximized. In the case where the signal from the SS is received in the BS at a higher power level, the performance of the SS is increased at the expense of increasing interference from other SSs sharing the same channel. As a result, system capacity is decreased or the call quality of other subscribers drops.
Orthogonal Frequency Division Multiplexing (OFDM)/Orthogonal Frequency Division Multiple Access (OFDMA) has recently been proposed as a physical layer scheme for a 4th generation mobile communication system. The above-described power control has also emerged as a challenging issue to the OFDM/OFDMA system.
OFDM/OFDMA is a transmission scheme based on the IEEE 802.16 standard, in which a serial modulation symbol sequence is transmitted as parallel data. OFDM/OFDMA operates in TDD. In OFDM, 256 modulation symbols are Fast-Fourier-Transformed (FFT-processed) to one OFDM symbol, whereas in OFDMA, one OFDM symbol is formed with more modulation symbols. According to the IEEE 802.16-based OFDMA, the subcarriers of one OFDM symbol are grouped into subchannels and a plurality of OFDM symbols form one frame.
FIG. 1 illustrates an OFDMA frame structure specified by IEEE 802.16. The horizontal axis represents OFDM symbol indexes and the vertical axis represents subchannel indexes.
Referring to FIG. 1, an OFDMA frame is comprised of a plurality of bursts each marked by a square on a time-frequency plane. Since the frame is time-division-duplexed, the downlink period and the uplink period can be flexibly controlled. For example, kth through (k+8)th symbols are allocated to the downlink and (k+9)th through (k+12)th symbols are allocated to the uplink, as illustrated in FIG. 1. In the OFDMA frame, a DL/UL MAP burst delivers configuration information (e.g. position, length, and MCS level) about a plurality of downlink/uplink bursts allocated to the frame. The bursts other than the DL/UL MAP burst transfer a DL/UL-Medium Access Control (MAC) layer control message and downlink/uplink data packets. Particularly, the control message bursts can be a power control change request/command message burst for controlling the power control scheme of each SS, or a power control message burst for controlling the transmit power of each SS. The bursts are time-division-multiple-accessed between SSs and the BS. Transmission gaps called transmit/receive transition gap (TTG) and receive/transmit transition gap (RTG) are inserted between the downlink and uplink periods.
Meanwhile, each SS performs initial ranging and periodic ranging to correct time and frequency errors in uplink bursts and control power. When the SS attempts ranging, the BS measures the power of a signal from the SS and transmits to the SS a MAC message including a compensation value for signal power loss caused by path attenuation and rapid signal power change.
Now a description will be made of an uplink power control method in a normal mode in the OFDM/OFDMA TDD system. The uplink power control is executed in two steps.
In the first step, the BS carries out power control. The BS scheduler determines available resources and an available MCS level for uplink transmission within the transmit power range of an SS of interest byΔP=SNRreq−SNRUL,RX+(BWreq−BWRX)+MARGINTX≦Headroom  (1)where SNRreq and BWreq respectively denote the required SNR and bandwidth for applying an MCS level to the current packet to be scheduled. SNRUL,RX and BWRX denote the received SNR and allocated bandwidth of a reference signal, respectively. The reference signal is a previously received uplink burst signal, a data signal or a control signal. MARGINTX is a term that represents a channel change. That, this margin is set considering the difference between the time of scheduling based on Equation (1) and the actual time of transmitting an uplink signal. Headroom is the transmit power margin of the SS, calculated by subtracting the current transmit power from the maximum transmit power of the SS. The BS is assumed to have knowledge of the maximum transmit power of the SS. ΔP satisfying Equation (1) ensures that the SS transmits an uplink signal with the resources and MCS level scheduled within the limited power.
In the second step, the SS performs power control. The uplink power control is considered in two ways: closed-loop power control and open-loop power control.
The uplink closed power control is a scheme of controlling the transmit power of the SS according to a command from the BS. The BS notifies the SS of a required power increment/decrement ΔP as well as the resources and MCS level scheduled by Equation (1).
The uplink open-loop power control is a scheme of deciding the uplink transmit power in the SS itself. The BS simply tells the SS the resources and MCS level decided by Equation (1) and the SS then computes the uplink transmit power of an uplink signal to be transmitted using the allocated resources by
                                                        P              =                            ⁢                                                PL                  UL                                +                                  SNR                  req                                +                                  NI                                      UL                    ,                    RX                                                  +                                  BW                  req                                +                                  MARGIN                  RX                                                                                                        =                            ⁢                                                PL                  DL                                +                                  SNR                  req                                +                                  NI                                      UL                    ,                    RX                                                  +                                  BW                  req                                +                                  MARGIN                  RX                                                                                                        =                            ⁢                                                PL                                      DL                    ,                    TX                                                  -                                  PL                                      DL                    ,                    RX                                                  +                                  SNR                  req                                +                                  NI                                      UL                    ,                    RX                                                  +                                  BW                  req                                +                                                                                                      ⁢                              MARGIN                RX                                                                        (        2        )            where PLUL and PLDL denote uplink and downlink path losses, respectively. In view of the TDD system, these two values are almost the same. The SS can estimate PLDL using the transmit power of the BS, PDL,TX and the downlink received power PDL,RX of the SS. NIUL,RX is the power of a signal and interference measured at a receiver of the BS, common to all of the SSs. SNRreq and BWreq respectively denote the required SNR and bandwidth for an MCS level to be applied to a packet. MARGINRX is a term that represents the difference between the time to which Equation (2) is computed for application and the actual uplink transmission time.
FIG. 2 is a diagram illustrating a signal flow for a conventional closed-loop power control.
Referring to FIG. 2, the SS transmits a reference signal and information about the uplink transmit power of the reference signal (UL_Tx, Power) in an uplink burst to the BS in step 201.
In step 203, the BS (scheduler) calculates the received SNR of the reference signal and determines resources, an MCS level, and a power increment ΔP for the SS by Equation (1). Headroom involved in Equation (1) can be calculated using the information of the transmit power (UL_Tx, Power).
In step 205, the BS allocates the uplink resources to the SS according to the scheduling (UL_MAP) and transmits a power control command (or the power increment) to the SS. The resource assignment (UL_MAP) information is delivered in a UL-MAP burst and the power control command is set in a DL burst containing a predetermined control message.
The SS determines its uplink transmit power according to the power control command in step 207 and transmits packets using the allocated resources in step 209. Thereafter, step 203 (BS scheduling) through step 209 (uplink transmission) are repeated.
As described before, the power control command is selectively transmitted in the closed-loop power control. Only if the channel status is changed and the SNR of an uplink received signal is changed, does the BS transmit a power control command to the SS. In the absence of the power control command, the SS determines its uplink transmit power based on the previous uplink transmit power byPnew=PLast+SNRNew−SNRLast+(BWNew−BWLast)  (3)where Pnew and PLast denote the new transmit power and the previous transmit power, respectively, SNRNew and SNRLast denote a required new SNR and the previous required SNR, respectively, and BWNew and BWLast denote a new allocated SNR and the previous allocated SNR, respectively.
FIG. 3 is a diagram illustrating a signal flow for a conventional open-loop power control.
Referring to FIG. 3, the SS transmits a reference signal and information about the uplink transmit power of the reference signal (UL_Tx, Power) in an uplink burst to the BS in step 301.
In step 303, the BS (scheduler) calculates the received SNR of the reference signal and determines resources, an MCS level, and a power increment ΔP for the SS by Equation (1). Headroom involved in Equation (1) can be calculated using the information of the transmit power (UL_Tx, Power).
In step 305, the BS allocates the uplink resources to the SS according to the scheduling (UL_MAP) and transmits the uplink resource assignment (UL_MAP) information to the SS. Compared to the closed-loop power control, a power control command is not transmitted in the open-loop power control. Instead, the BS broadcasts in a DL-MAP burst PDL,TX and NIUL,RX necessary for the computation of Equation (2) to all of the SSs.
The SS determines its uplink transmit power using the resource assignment information by Equation (2) in step 307 and transmits an uplink signal using the allocated resources in step 309. At the same time, the SS tells the BS the current transmit power. Thereafter, step 303 (BS scheduling) through step 309 (uplink transmission) are repeated.
As described earlier, in contrast to the closed-loop power control, the open-loop power control scheme provide to the BS information about the current uplink transmit power along with the uplink transmission because the SS can change the uplink transmit power freely. Equation (2) that the SS uses in deciding the transmit power includes a channel variation which is not known to the BS and thus the headroom of the SS is changed, unnoticed by the BS. Therefore, the SS tells the BS the current transmit power at every uplink transmission so that the BS can update the headroom.
On the other hand, in the closed-loop power control, the transmit power of the SS is changed by a power control command from the BS or a transmit power calculation formula (Equation (3)) known to the BS. Accordingly, the BS can distinguish a transmit power change from a channel change in the SNR estimate of an uplink signal. That is, the BS can execute a power control taking the channel change into account, as shown in Equation (1). The headroom can also be calculated using the previous headroom and the previous power control command or using the transmit power of the SS that the bas station can estimate by Equation (3). Consequently, the SS does not need to notify the BS of its transmit power at every uplink transmission in the closed-loop power control.
The features of the two power control schemes are summarized below in Table 1.
TABLE 1Closed-looppower controlOpen-loop power controlDownlink feedbackPower controlPDL,TX, NIUL,RXcommandUplink feedbacknoneUplink transmit powerScheduling marginMARGINTXMARGINTXMaximum transmit powerMARGINTXMARGINRXmargin
As noted from Table 1, the closed-loop and open-loop power control schemes differ in uplink/downlink feedback, scheduling margin, and maximum transmit power margin. The uplink/downlink feedback has been described before. The scheduling margin is MARGINTX in both power control schemes because a scheduling time point coincides with an actual uplink transmission time in them. The maximum transmit power margin is defined as the maximum difference between a required transmit power satisfying SNRreq at the receiver and an actual transmit power. For the closed-loop power control, the maximum transmit power margin is MARGINTX since the actual transmit power is decided at scheduling. For the open-loop power control, the actual transmit power is decided by Equation (2) and thus the maximum transmit power margin is MARGINRX. The scheduling margin leads to resource assignment loss, and the maximum transmit power margin results in an increase in total system interference.
If the SS moves slowly, the closed-loop power control performs better on the whole. Because the channel does not change much at a low mobile velocity, the power control command is not issued frequently and thus the amount of downlink feedback information is small. MARGINTX affected by the channel variation is also very small. Also, the scheduling is done and the transmit power is decided according to the actual uplink channel status, as in Equation (1). Therefore, the uplink power control can be performed with high reliability.
On the contrary, if the SS moves fast, the open-loop power control outperforms the closed-loop power control. The channel changes greatly at a high mobile velocity and thus the number of occurrences of the power control command in the closed-loop power control is approximately equal to the number of transmit power feedbacks in the open-loop power control. However, because MARGINTX MARGINRX, the closed-loop power control tracks the channel variation consuming much resources, or cannot track the channel variation at all. As a result, the closed-loop power control causes greater interference than the open-loop power control in the case where the SS moves fast.